Heatmaps and behaviour analysis

Visualise how users actually interact with your website to make better decisions

9 min

Traditional analytics tells you what happens (bounce rates, drop-off, conversions) but not why. Heatmaps and session recordings fill that gap: they show exactly where users click, how far they scroll, what confuses them and what they ignore.

This guide explains the types of heatmaps, how to interpret session recordings, which tools to use and how to turn these visual analyses into concrete optimisation hypotheses.

Types of heatmaps

Heatmaps represent interaction data as colour gradients overlaid on the page: red zones (high activity) and cold zones (low activity). Each type of heatmap answers a different question about user behaviour.

  • Click maps: where users click. Reveal whether CTAs get attention or if users click on non-clickable elements (false affordance)
  • Scroll maps: how far users scroll. Show what percentage of users see each section of the page
  • Move maps (hover maps): where users move their cursor. Partial correlation with visual attention, useful on desktop
  • Attention maps: weighted combination of time, scroll and clicks to estimate actual attention

How to interpret heatmaps correctly

A heatmap is not an answer: it is a clue. Seeing that 60% of users never reach the main CTA is valuable data, but the solution could be moving the CTA up, reducing the preceding content or making the value proposition more compelling to motivate scrolling. The heatmap identifies the problem; the hypothesis proposes the solution.

Watch out for the volume trap: in a click map, navigation menu areas will always be the reddest because everyone uses them. That does not mean they are the most important. Filter by relevant segments: converting users vs non-converting, mobile vs desktop, new vs returning.

Session recordings

Session recordings replay a user’s complete interaction: cursor movements, clicks, scrolling, page changes, errors. They are the most empathetic tool in digital analytics: you see the website through the user’s eyes.

Do not try to watch every recording: it is impossible and impractical. Filter by relevant sessions: users who abandoned checkout, users who visited the pricing page but did not convert, or users who spent over 3 minutes on a form without submitting. Look for patterns, not anecdotes.

  • Filter by behaviour: abandonment, rage clicks, error pages
  • Look for recurring patterns, not individual anomalous sessions
  • Combine with analytics data to contextualise what you see
  • Document insights with screenshots and timestamps for the team

Rage clicks, dead clicks and UX errors

Modern tools automatically detect frustration signals. Rage clicks are rapid repeated clicks on the same spot: they indicate something is not working or not responding as the user expects. Dead clicks are clicks on non-interactive elements: the user believes it is a link or button, but it is not.

These frustration signals are gold for CRO: they pinpoint exactly where the experience fails. A high volume of rage clicks on the "apply coupon" button may reveal a bug that is costing thousands in lost sales.

Behaviour analysis tools

Hotjar is the most popular tool due to its ease of use and freemium model. It offers heatmaps, recordings, surveys and feedback widgets in a single platform. Microsoft Clarity is completely free and includes good-quality heatmaps and recordings.

FullStory and Contentsquare are enterprise options with advanced capabilities: complete journey analysis, automatic UX error detection, granular segmentation and quantification of each detected issue’s conversion impact.

  • Hotjar: most popular, freemium, heatmaps + recordings + surveys
  • Microsoft Clarity: completely free, good quality, GA4 integration
  • FullStory: enterprise, journey analysis, automatic error detection
  • Contentsquare: enterprise, conversion impact quantification

From insight to action: the full process

Behaviour analysis generates hypotheses, not decisions. The full process is: collect heatmap and recording data → identify problem patterns → formulate hypotheses using the "If-Then-Because" format → prioritise with ICE → test with A/B testing → implement winners.

Document each insight with visual evidence (heatmap screenshots, recording clips) and connect to quantitative GA4 data. An insight stating "users do not see the CTA" accompanied by a scroll map where only 30% reach it and GA4 data showing 85% bounce on that page is far more convincing than an opinion.

Key Takeaways

  • Heatmaps show the what; session recordings show the why
  • Always filter and segment: aggregated data hides key patterns
  • Rage clicks and dead clicks pinpoint exact frustration points
  • Hotjar and Microsoft Clarity cover most teams’ needs
  • Behaviour insights should become testable hypotheses, not direct decisions

Want to understand how users behave on your website?

We implement behaviour analysis tools, identify improvement opportunities and design tests to optimise experience and conversion.